4D Machine Installation Using Navisworks

JOE KAWKA: Hello. My name is Joe Kawka. I'm a digital engineer at Toyota Motor North America. And today I'm going over how to make a 4D machine installation using Navisworks. And essentially, we're building an installation. And this comes about because our installations are very large and hard to handle. So Navisworks is a great tool to utilize to really organize everything. And this is the Safe Harbor Statement.

So before we jump into it, what I really want to overview is what was our previous method for these really large volume installations. There are many instances where we're handling hundreds of machines at a time, not just in powertrain, which is the division I'm in, but also body weld, stamping, paint, and so on-- any of our divisions, really. And our methods can vary from department to department, but this is my method that I am implementing with powertrain specifically, and hope to share to you guys also.

And so, previously, we would have to cross reference many, many documents, many documents. So 2D layouts often from an uncontrolled drive, we would have to print them out. We would need large screens to review them. We would have very large meetings to go over them, and really get eyes on the presentation and the installation itself.

But also, it was very difficult to find specific assets and their locations based on our schedule. And also, our schedule, we have a proprietary in-house built scheduler that we've made. And it does get very long. I've seen schedules that span the very wall from ceiling to floor. And it's very large, and convoluted, and hard to follow. But that's something we have to deal with.

And also, in addition to that, every machine has many phases to its installation. Not only are we tracking it from arrival, but sending it down, hooking it up to electric, any other utilities that it needs, all the set, the adjustments, and, finally, hand over to whoever needs the machine. And also, there are many different types of assets that we need to manage within it. So not just machines, but stations, conveyors, AGVs, and so on.

And lastly, we have specification drawings that are basically very detailed drawings of each machine that we get from our vendors. And when we're referencing them, where it's very difficult to read, just because of the nature of there's a lot of detail about this machine that we definitely need, but not necessarily for the installation. And we only really reference it when the areas get tight, but that really relies on the expertise of our team members to know what areas are tight. And many times, just because of human error, that could be missed fairly easily.

And so with that in mind, all this cross-referencing from one document to another and having physical copies going from one to the other, there are many opportunities for many mistakes. And every mistake, it comes at a cost. Field orders are costly in both time and money. They can set back our installation timing. That sets back a whole cascade of other processes down the line. And we definitely don't want to do that.

Also, if we do want to try to pull up that activity and make sure that we're doing our due diligence of cross-referencing everything, that, too, is going to take a lot of the engineers' time to review, to have meetings, and just thinking about it. It's a lot. And by nature of our processes, because we have so many machines in a row, a small issue can spiral out of control very quickly. For instance, if we need to move a machine because we've realized that one of our columns are off by a few hundred millimeters, then there is a very good potential that every machine that's linked to it, which can be up to 20 or 30 machines in a row, they would also need to move relative to that one machine. And then in addition to that, utilities have to move. And as you can see, as you could think, it cascades. It exponentially grows, it seems.

And also, I do want to point out that time to cross reference all three forms of the document can take weeks. If we really want to make sure we don't have any mistakes, it's very time consuming. And a lot of manual effort has to go into it to make sure that actually happens correctly.

And so we have these issues. We've had these issues. So what's the solution? How do we reduce these errors. How do we reduce the time to reference from one document to another document? How do we get from the 2D and our schedule and the machine itself? And how do we combine those all together? And that's where Navisworks can really thrive. And that's where my solution really came into fruition for my team.

To link it all together, I would like to start with the factory design utilities. So to link it in with the bottom area here and the 3D assets, factory design utilities is going to be the best tool to have that flexibility to work both in 2D and 3D layouts, where you have your old school guys that love their 2D layouts, and then 3D layouts that are really encompassing all of the information that you need, especially when it comes to rigging and installation.

From there, the scheduler, we have our own scheduler. And we've kind of customized our output as a CSV. But there are many project scheduling softwares out there that do export in CSV. And you'll see later on that Navisworks will be able to handle these various CSV files and formats.

And then, lastly, I like to point out how I'm linking it all together. Both factory design utilities and scheduler feed into the simulation. And that's Navisworks. Navisworks, within there, we would be using TimeLiner, Animator, Clash Detective, all these functions that are all included in there. And lastly, Navisworks would be able to either report it out, or have our own internal way of reporting out that you guys might like also.

And so to start with the baseline, we needed to create these 3D asset libraries. And the best practice is to have it all in the Vault, which it can be easily referenced especially in Inventor, and also well controlled, which is a big downfall of a shared drive. And also, with creating the FDU, or Factory Design Utilities, like I mentioned before, you get the best of both worlds. You have the lightweightness of the 2D layouts in AutoCAD. And then you have Inventor and 3D. And you can really fine tune all your layouts. In addition, it's linked directly to the Navisworks simulation model, meaning any 2D updates you make, it can reflect immediately, or almost immediately, to the Navisworks file. So there is no confusion about if this is the latest one or not.

And lastly, the schedule, mostly matching the categories within Navisworks TimeLiner. And that's the start date, end date, all your different categories of what kind of phase your machine is in, so setting, adjusting, turning on, what have you. Those can all be within TimeLiner.

And so now we have our baseline. But how do we combine our schedule, our CSV file with Navisworks? And I'll go into a more technical dive into how that's done exactly. So adding in the CSV file into TimeLiner, you would go to Data Sources in the TimeLiner window. Hit Add. Select the CSV, and a pop-up will show up with Field Selector.

And from there, with your CSV, you're going to match your columns in the Navisworks with your external fields. So in your CSV file, you'll have columns of task type, external ID. You can see here actual start date, actual end date. And that's going to line up directly with, again, the Navisworks task type synchronization ID, actual start date, actual end date. So the diction might be a little bit different, but that's not an issue, as long as you create these links appropriately.

And from there, we're going to load or refresh the newly added CSV, or refresh if you already have one in there, by quite simply going into Refresh from Data Source, Rebuild Task Hierarchy, OK, and that's for the first time. For anything new, you could use Synchronize. So we'll say your project schedule is updating daily, and you need to just run these simulations every day. It's as quick as updating the CSV in this Navisworks file, and then rerunning your simulation. And you're good to go. You're good to start sharing it with people.

And as you can see, we're retrieving the data from the CSV for the columns that we specified in the last slide. And then we're exporting these sets. And what that does, it creates all these sets so that we can start linking the individual assets to the sets or the assets that we have in our CSV, just helping us with creating that connection really.

And so we're linking the machines to sets. You're going to need both the selection tree and the sets window open. You can go in the View tab, Windows, and just make sure those two are checked. And in the topmost level of the machine, so meaning in the tree here, just make sure that it's not expanded. Just pick the machine that you want in the very top. And select that machine, pan over to the set portion, right-click on the appropriate asset that it's representing, and then click Update. And that creates that link from the 3D model to the sets.

And then you're going to have to repeat this. Thankfully, this is only needed to be done once when you set it up. Every time you update the simulation with the CSV, you will not need to recreate these links. But unfortunately, you do have to make these links for every machine to start out.

And then, like I kind of mentioned before, we exported the sets out of the TimeLiner, but now we're going to bring it back into the TimeLiner and create that link again. But this time, we're going to use an auto-attach rule. So that really automates things and speeds things up as far as the setup. Again, you don't need to do it after this is already set up. But you can see here, Auto-Attach using Rules, Create New Rule, Attach Items to Task. Automatically, when you click on this number three portion, number four will automatically pop up.

But this diction might be a little bit different. In my case, items came up first, and so I had to change it to selection sets. From there, OK. Make sure the new rule is checked. And then Apply Rules. From there, all the connections should be made immediately.

And once that's done, you can add in each event. So in my case, we have different acronyms for different phases of the machines, but pretty straightforward-- construction, demolished, temporary already in there. But you can add as many as you want as far as setting in adjust, power on, or hand over.

And so this is where you can really configure what the visual is going to be of each machine in both color and transparency for each phase that the machine will pass through. And we'll see an example of this later on.

And then, lastly, when we are refreshing, like I mentioned before, with the CSV, you can go to Data Sources tab. But instead, this time, choose Synchronize. And that's after all of this new data that we've linked together. Synchronize will ensure that we're moving forward with all the right data in the right spots, really finalizing those links.

So that's concluding the TimeLiner portion of it. And now like, how do we make an animation? We have this machine that's sitting in our yard, in this tent, and we have our riggers ready to go. But we want to make sure that these machines are able to fit through our doors, which seems to be a big problem with many, many projects. And how do we incorporate animation into this TimeLiner? And so how do we move it in Navisworks?

And so to start off, we will add a scene. Obviously, you have to open up the Animator tab. And then it's pretty straightforward. We will press the green plus sign and add a scene. Right-click on the newly created scene, and go to Add Animation Set. And from there, you can add whatever asset that you would like to. So that's just setting up the animation for it to really specify what you want to do with it.

And so now that we've added the scene, we can add the machine. So we'll select the machine that we want to animate from the Sets tab. So to do that, we'll right-click on the camera scene. We'll hit the camera. Or hint, the camera is not needed, but can be useful to follow a particular path.

So what I like to do, personally, is to not have a camera. And that allows me to really adjust my views on the layout, and really drill in into an area if I need to. But a camera can be useful if you need to follow a particular machine. And this is where you would set it up. And you would set it up the same way with viewport to viewport, as you would a machine. And we'll get into that in a second here. But anyways, you could either choose from current selection or from current search or selection set.

So when we have our machine set, you can see there's a row. In our case, it's a JTech tote carrying machine. And there are no diamonds, meaning there is no snapshot of where it's supposed to be. And we also want to make sure, just to start, that it is the active one. So we've selected it.

And then we want to put either-- it's weird. It's more up to personal preference, if you want to start at your set position, or if you want to end at your set position. Personally, it makes sense to me to have your machine set at your end position. / you can put an arbitrary number. In this case, we're putting in 10 seconds. And we're setting the position at that time and capturing that key frame. And the Save button to save the machine at the set position at 10 seconds, meaning when the animator gets to 10 seconds, the machine will be in its final resting place.

I think another key note is once we bring this animator into the TimeLiner, the span of time, how many seconds you have this machine moving from one spot to another, is almost arbitrary. So there is no need to really focus on how much time, how fast does this machine travel through our facility to get to its final resting place. Again, like I said, when it's at the resting place or the final set place at 10 seconds, capture key frame, and a diamond will appear on the set's row at the time specified in this case.

And so now to move it, there is a move icon, a rotate, and a scale. In my case, I'm never going to use the scale, but you never know, but mostly, translate and rotate. So the sequence that I use is adjusting the time of the animator. You'll see the vertical line move to whatever time you specify. I think it's the black line.

And then you would select either translate or rotate. And you would move the machine using the gimbal, I believe, is what they call it. And then once that is in a position that you would like, then you capture the key frame. And so you would repeat this over and over until you're satisfied with the machine's path, if you would. There's also fine-- once you select on the translator or the rotating options, you can fine tune with however much accuracy you want to move the machine, or put in a numerical value into where you need.

And then, so once our-- so now we have our TimeLiner done. And now we have our animation done. So how do we bring those two together? And that's linking the animation per asset here. So back in the Timeline tab, we need to right-click on the-- I'm sorry-- we need to select the column dropdown, and then extend. And this is because the default setting for the TimeLiner is to hide the specific column that we need to activate.

So once that's extended, we can scroll over to Animation and select the specific animation associated with the asset, meaning-- I would suggest labeling a lot better than I have, where you have your asset number and your animation number match, or at least know the linkage between those two. And this is where I actually create that linkage.

And then, also, like I mentioned before, the animation behavior, scaling it goes from actual start to actual end. And it adjusts the amount of time it takes for that machine to travel that whole animation. If you do want it to be more proportional, you have the option to. But I find it more difficult than helpful.

And so now we have our TimeLiner. That represents a schedule. We have our animation and our layout. That represents what are we physically going to do. And now we need to solve like our crashes. These are money pits. Are we going to run into utilities that we are trying to install ahead of time? Are we going to run into other machines that we're-- because we're delayed on a particular machine? So how do we detect these crashes in Navisworks?

And the solution is pretty straightforward. If you've ever used Clash Detective before, it's as simple as adding in the link to TimeLiner. And we'll go into TimeLiner-- or I'm sorry-- we'll go into Clash Detective tab, Settings, Link. Go to the dropdown, and select TimeLiner. I think something to really note, that I may have needed to restart many simulations on is that the default step time is 0.1 of a second, meaning it will stop the simulation at 0.1 of a second, do all its Clash Detective stuff in the background, and then move another 0.1 of a second, and then do it again.

And that is unfathomable of how long it would take for our projects, because we're spending months. So I suggest bumping it up by many timescales. I think either by hours, days-- or hours, minutes, probably not too large, just to think about the installation paths and making sure that you're catching most, if not all, of your collisions. Just a warning.

And then, so now with Clash Detective, obviously, it spits out all the results, which is great. But now we have a kind of a separation of, what are the static clashes by design of our layout? What are the issues that we can fix, that we need to fix, as far as like moving machines around? Or is it something that's dynamic, where the machines are moving and they're clashing into each other as they're moving, but in their final resting spot they're just fine?

So to sort through those two things, you would go to in the Results tab the Events, and Choose Columns, and Events is checked. The animated clashes will have an event tied to the name. And it's the same name as the TimeLiner task. And then the static clashes or the clashes that are just naturally there from the layout will be blank under that column. And you can see, in this case, this is two moving machines running into each other. And it's showing that it's happening during an event.

And so now we have all this great information. We have our 3D layout. We have our schedule. We have those combined. We have our animation. We have our clashes. We have those combined, and we have this nice simulation with all these issues ready to be resolved.

But how do we tell people that we found all of these? Or even, what's our plan from day to day? Because we're going months with hundreds of machines. And every day, we're either installing a machine. We're adjusting 10 machines. We're powering on 20 other machines. How do we make sure that everybody's on schedule in a way that is irrefutable, that nobody can mess up? Like, I thought I was working on that machine, and not the one next to it. And this is, how do we collaborate? How do we get that information out there?

Luckily, TimeLiner has an export of pictures. And that's what we use internally to this day. And once we are done simulating in the TimeLiner, and this is more focused solely for the installation, less so for the clashes, but still it could be used. But once we get the simulation of the TimeLiner done, we're going to export a picture of the layout per day. And we're spinning it through our whole installation. And you can see our installations are 141 days. And we're just specifying different views that we want to see. And you'll see an example in a minute here.

Also, exporting animation in TimeLiner is a possibility. So you can see the machine's actually moving. And it's, again, you do frames per second. I believe you do an MP4 video also. And it will export out.

In our case, we have made a PowerPoint macro, essentially AddInsInstaller.vbs, our and then a save images PPAM file. I can probably get some more information if needed. But long story short, what it does, it takes all of the images that Navisworks has spit out and overlays it into a PowerPoint that anybody and everybody can open up, that most people can see, I would hope. And you can see every day what are the changes, what machines are in what state, and so on.

So this is an example, not an actual case, but this is something that we had outputted to our team to aid in installation. So this is a great tool to really communicate out to the field to a mass amount of people, indicating I want these specific machines to be in adjustment on 12/2/2022. Or it gives another opportunity for people to be prepared, like, OK, this machine in the middle of all these others that are further advanced, this is still in the power on phase. We need to adjust our schedule accordingly to bring everybody up at the same time. So maybe that's a-- it's a good visual and quick representation and communication out to many experts, many people that can raise their hand, bring up issues as needed, or whatnot.

And then, as I mentioned before, the TimeLiner is able to export videos. In this case, you can see that the camera is changing as the animation goes along. And this is a part of the following, the specific machinery capability that you can bring in if you need. In this case, we're able to identify clearly this box is not able to fit through the front doors.

But the rest of it is able to transport just fine. So maybe we just need to figure out a workaround and rigging to get it through the doors in the first place. And this is an example of being able to identify issues before you have millions of dollars of machines sitting on the ground, waiting for it to be installed, or brought into the building with inclement weather.

So lastly, Navisworks naturally does have an output of clash reports. So this is very helpful when it comes to resolving issues, both in installation and the layout itself. You can see, in this case, we have 196. And that could have been 196 field orders that have now been identified before even a machine has showed up at our facility. And we're able to measure these preemptively, saving us lots of headache and cost. I think it's a very useful tool that can show both of the aspects of the installation and the final version itself.

And I believe that is it. Thank you for listening in. I hope you have learned something new. Feel free to reach out to me with any questions. I'd be more than happy to help. Anything else? Have a great day. Thank you.